Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

The reference human genome sequence set the stage for studies of genetic variation and its association with human disease, but epigenomic studies lack a similar reference. To address this need, the NIH Roadmap Epigenomics Consortium generated the largest collection so far of human epigenomes for primary cells and tissues. Here we describe the integrative analysis of 111 reference human epigenomes generated as part of the programme, profiled for histone modification patterns, DNA accessibility, DNA methylation and RNA expression. We establish global maps of regulatory elements, define regulatory modules of coordinated activity, and their likely activators and repressors. We show that disease- and trait-associated genetic variants are enriched in tissue-specific epigenomic marks, revealing biologically relevant cell types for diverse human traits, and providing a resource for interpreting the molecular basis of human disease. Our results demonstrate the central role of epigenomic information for understanding gene regulation, cellular differentiation and human disease.

Integrative analysis of 111 reference human epigenomes

Persistent postsurgical pain: risk factors and prevention

The brain is the key organ of the response to stress because it determines what is threatening and, therefore, potentially stressful, as well as the physiological and behavioral responses which can be either adaptive or damaging. Stress involves two-way communication between the brain and the cardiovascular, immune, and other systems via neural and endocrine mechanisms. Beyond the "flight-or-fight" response to acute stress, there are events in daily life that produce a type of chronic stress and lead over time to wear and tear on the body ("allostatic load"). Yet, hormones associated with stress protect the body in the short-run and promote adaptation ("allostasis"). The brain is a target of stress, and the hippocampus was the first brain region, besides the hypothalamus, to be recognized as a target of glucocorticoids. Stress and stress hormones produce both adaptive and maladaptive effects on this brain region throughout the life course. Early life events influence life-long patterns of emotionality and stress responsiveness and alter the rate of brain and body aging. The hippocampus, amygdala, and prefrontal cortex undergo stress-induced structural remodeling, which alters behavioral and physiological responses. As an adjunct to pharmaceutical therapy, social and behavioral interventions such as regular physical activity and social support reduce the chronic stress burden and benefit brain and body health and resilience.

/pdf/physiology-and-neurobiology-of-stress-and-adaptation-central-2xtqi7pnfe.pdf

Physiology and Neurobiology of Stress and Adaptation: Central Role of the Brain

A neurotrophic model for stress-related mood disorders.

Anxiety and stress response/resiliency are heritable traits central to the etiology of multiple psychiatric diseases, but efforts to identify genetic variation influencing this broad domain of neurobiological function are hampered by the coarseness of the phenotypic measures and the effects of environmental factors. Neuroimaging offers a powerful approach for assessing functional neuronal activity. Neurophysiological measures can serve as intermediate phenotypes more directly linked to small gene effects, compared with behavioral end points of neural dysfunction. Imaging genomics is a relatively new research area that is concerned with linking functional gene variants and brain information processing. Here, we will focus on processes affected by anxiety and stress. Neuroimaging has been combined with genetic analysis to reveal genetic effects of functional variants of the serotonin transporter (5-HTT) and catechol-O-methyltransferase (COMT) genes on brain response to stressful stimuli. The low-expressing allele of the 5-HTT promoter polymorphism (HTTLPR) is associated with anxiety and with greater amygdala and other regional responses to emotional. The COMT Met158 allele leads to lower COMT activity and has also been associated with anxiety, and the effect of this gene is apparently additive with HTTLPR. Individuals with Met158 genotypes are more sensitive to pain stress and, as shown by C11 Carfentanil imaging, have diminished ability to upregulate opioid release after pain/stress. These results suggest that functional variants of 5-HTT and COMT impact brain functions involved in stress and anxiety.

Imaging genomics applied to anxiety, stress response, and resiliency

Intravenous illicit drug use (IDU) and hepatitis C infection (HCV) commonly co-occur among HIV-infected individuals. These co-occurring conditions may produce interacting epigenetic effects in white blood cells that influence immune function and health outcomes. Here, we report an epigenome-wide association analysis comparing IDU+/ HCV+ and IDU-/HCV- in 386 HIV-infected individuals as a discovery sample and in 412 individuals as a replication sample. We observe 6 significant CpGs in the promoters of 4 genes, NLRC5, TRIM69, CX3CR1, and BCL9, in the discovery sample and in meta-analysis. We identify 19 differentially methylated regions on chromosome 6 harboring MHC gene clusters. Importantly, a panel of IDU+/HCV+-associated CpGs discriminated HIV frailty based upon a validated index with an area under the curve of 79.3% for high frailty and 82.3% for low frailty. These findings suggest that IDU and HCV involve epigenetic programming and that their associated methylation signatures discriminate HIV pathophysiologic frailty.

/pdf/dna-methylation-signatures-of-illicit-drug-injection-and-yv59ab82ei.pdf

DNA methylation signatures of illicit drug injection and hepatitis C are associated with HIV frailty

Epigenome-wide association analysis revealed that SOCS3 methylation influences the effect of cumulative stress on obesity.

The effects of tobacco smoking on epigenome-wide methylation signatures in white blood cells (WBCs) collected from persons living with HIV may have important implications for their immune-related outcomes, including frailty and mortality. The application of a machine learning approach to the analysis of CpG methylation in the epigenome enables the selection of phenotypically relevant features from high-dimensional data. Using this approach, we now report that a set of smoking-associated DNA-methylated CpGs predicts HIV prognosis and mortality in an HIV-positive veteran population. We first identified 137 epigenome-wide significant CpGs for smoking in WBCs from 1137 HIV-positive individuals (p < 1.70E−07). To examine whether smoking-associated CpGs were predictive of HIV frailty and mortality, we applied ensemble-based machine learning to build a model in a training sample employing 408,583 CpGs. A set of 698 CpGs was selected and predictive of high HIV frailty in a testing sample [(area under curve (AUC) = 0.73, 95%CI 0.63~0.83)] and was replicated in an independent sample [(AUC = 0.78, 95%CI 0.73~0.83)]. We further found an association of a DNA methylation index constructed from the 698 CpGs that were associated with a 5-year survival rate [HR = 1.46; 95%CI 1.06~2.02, p = 0.02]. Interestingly, the 698 CpGs located on 445 genes were enriched on the integrin signaling pathway (p = 9.55E−05, false discovery rate = 0.036), which is responsible for the regulation of the cell cycle, differentiation, and adhesion. We demonstrated that smoking-associated DNA methylation features in white blood cells predict HIV infection-related clinical outcomes in a population living with HIV.

/pdf/machine-learning-selected-smoking-associated-dna-methylation-5c1r67zmng.pdf

Ke Xu

Papers

Imaging genomics applied to anxiety, stress response, and resiliency

DNA methylation signatures of illicit drug injection and hepatitis C are associated with HIV frailty

Epigenome-wide association analysis revealed that SOCS3 methylation influences the effect of cumulative stress on obesity.

Machine learning selected smoking-associated DNA methylation signatures that predict HIV prognosis and mortality

Similar psychotic and cognitive profile between ketamine dependence with persistent psychosis and schizophrenia.